Add warning for deprecated to_edge() flow with CoreML #15963

Lokesh9106 · 2025-11-24T18:35:38Z

This commit adds a warning message in the CoreML partitioner's partition() method to alert users when they use the older to_edge() flow instead of the recommended to_edge_transform_and_lower() flow.

The warning informs users that:

Using the old flow may result in performance regression
The recommended approach is to use to_edge_transform_and_lower() with CoreML partitioner
Provides a link to the CoreML backend documentation for more details

Fixes #15960

This commit adds a warning message in the CoreML partitioner's partition() method to alert users when they use the older to_edge() flow instead of the recommended to_edge_transform_and_lower() flow. The warning informs users that: - Using the old flow may result in performance regression - The recommended approach is to use to_edge_transform_and_lower() with CoreML partitioner - Provides a link to the CoreML backend documentation for more details Fixes pytorch#15960

pytorch-bot · 2025-11-24T18:35:43Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15963

📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 6 New Failures

As of commit 7f967ac with merge base 93bf861 ():

NEW FAILURES - The following jobs have failed:

Build Presets / windows (windows) / build (gh)
Lint / lintrunner / linux-job (gh)
>>> Lint for backends/apple/coreml/test/test_coreml_partitioner.py:
pull / unittest / linux / linux-job (gh)
export/tests/test_target_recipes.py::TestTargetRecipes::test_w2l_model
pull / unittest / macos / macos-job (gh)
export/tests/test_target_recipes.py::TestTargetRecipes::test_w2l_model
pull / unittest-editable / linux / linux-job (gh)
export/tests/test_target_recipes.py::TestTargetRecipes::test_w2l_model
pull / unittest-editable / macos / macos-job (gh)
export/tests/test_target_recipes.py::TestTargetRecipes::test_w2l_model

This comment was automatically generated by Dr. CI and updates every 15 minutes.

meta-cla · 2025-11-24T18:35:46Z

Hi @Lokesh9106!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at [email protected]. Thanks!

github-actions · 2025-11-24T18:36:13Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

meta-cla · 2025-11-24T19:09:30Z

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

metascroy · 2025-11-24T19:28:43Z

Great start! But I'm afraid adding a warning inside the partitioner itself isn't sufficient because the partitioner is also used in the to_edge_lower_and_transform flow.

Perhaps you could take a look at the similar change for XNNPACK: #13209

We should also add some unit tests showing: 1) the warning is logged with to_edge, but 2) not logged with to_edge_lower_and_transform.

Lokesh9106 · 2025-11-24T19:33:43Z

Thank you for your appreciation Can I make changes and resubmit it Get Outlook for Android<https://aka.ms/AAb9ysg>

…

________________________________ From: Scott Roy ***@***.***> Sent: Tuesday, November 25, 2025 12:59:05 AM To: pytorch/executorch ***@***.***> Cc: BELLAM LOKESH ***@***.***>; Mention ***@***.***> Subject: Re: [pytorch/executorch] Add warning for deprecated to_edge() flow with CoreML (PR #15963) [https://avatars.githubusercontent.com/u/161522778?s=20&v=4]metascroy left a comment (pytorch/executorch#15963)<#15963 (comment)> Great start! But I'm afraid adding a warning inside the partitioner itself isn't sufficient because the partitioner is also used in the to_edge_lower_and_transform flow. Perhaps you could take a look at the similar change for XNNPACK: #13209<#13209> We should also add some unit tests showing: 1) the warning is logged with to_edge, but 2) not logged with to_edge_lower_and_transform. — Reply to this email directly, view it on GitHub<#15963 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BXDRGNQCC4SIIILM7WQWYDL36NMADAVCNFSM6AAAAACNB2NXYGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTKNZSGM4DQMRRGU>. You are receiving this because you were mentioned.Message ID: ***@***.***>

metascroy · 2025-11-24T20:02:26Z

Can I make changes and resubmit it

Absolutely!

… workflow Following the pattern from PR pytorch#13209 (XNNPACK), this commit: 1. Adds `import inspect` to enable call stack inspection 2. Adds `_check_if_called_from_to_backend()` helper method that: - Returns False if called from to_edge_transform_and_lower (recommended flow) - Returns True if called from to_backend in deprecated flow 3. Wraps the deprecation warning in a conditional check This ensures the warning only appears when using the deprecated to_edge() + to_backend() workflow, not when using the recommended to_edge_transform_and_lower() flow. Unit tests will be added in a separate commit.

Adds two unit tests to verify the deprecation warning behavior: 1. `test_deprecation_warning_for_to_backend_workflow`: Verifies that the warning IS logged when using the deprecated to_edge() + to_backend() workflow 2. `test_no_warning_for_to_edge_transform_and_lower_workflow`: Verifies that the warning is NOT logged when using the recommended to_edge_transform_and_lower() workflow Also adds necessary imports: - import io - import logging - from executorch.exir import to_edge, to_edge_transform_and_lower These tests ensure the deprecation warning only appears in the deprecated flow, following the same pattern as XNNPACK (PR pytorch#13209).

metascroy · 2025-11-25T18:07:15Z

Looks great! Make you you run the linter though.

To run the linter, execute the following from the executorch repo folder:

lintrunner -a

This requires you install the lintrunner (pip install -r requirements-lintrunner.txt).

metascroy · 2025-11-25T18:09:08Z

backends/apple/coreml/test/test_coreml_partitioner.py

    # test_runner.test_symint_arg()
    test_runner.test_take_over_constant_data_false()
+
+        def test_deprecation_warning_for_to_backend_workflow(self):


Shouldn't these be methods inside TestCoreMLPartitioner?

Then in the main section you can add test_runner.test_deprecation_warning_for_to_backend_workflow() and test_runner.test_no_warning_for_to_edge_transform_and_lower_workflow()

metascroy · 2025-11-25T18:12:06Z

backends/apple/coreml/partition/coreml_partitioner.py

            ), "When lower_full_graph=True, you must set take_over_mutable_buffer=True"

+
+            def _check_if_called_from_to_backend(self) -> bool:


I think the indentation here is off. As written, this is defined inside init

pytorch-bot · 2025-12-03T18:56:08Z

To add the ciflow label ciflow/trunk please first approve the workflows that are awaiting approval (scroll to the bottom of this page).

This helps ensure we don't trigger CI on this PR until it is actually authorized to do so. Please ping one of the reviewers if you do not have access to approve and run workflows.

metascroy · 2025-12-03T18:56:23Z

Are you still working on this @Lokesh9106 ?

Lokesh9106 requested review from cccclai and shoumikhin as code owners November 24, 2025 18:35

meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Nov 24, 2025

Lokesh9106 and others added 3 commits November 25, 2025 08:49

Merge branch 'main' into Lokesh9106-patch-1

4f121fd

metascroy reviewed Nov 25, 2025

View reviewed changes

metascroy added 2 commits December 1, 2025 11:50

Merge branch 'main' into Lokesh9106-patch-1

5e2a46b

Merge branch 'main' into Lokesh9106-patch-1

7f967ac

metascroy added the ciflow/trunk label Dec 3, 2025

pytorch-bot bot removed the ciflow/trunk label Dec 3, 2025

		), "When lower_full_graph=True, you must set take_over_mutable_buffer=True"


		def _check_if_called_from_to_backend(self) -> bool:

Add warning for deprecated to_edge() flow with CoreML #15963

Are you sure you want to change the base?

Add warning for deprecated to_edge() flow with CoreML #15963

Uh oh!

Conversation

Lokesh9106 commented Nov 24, 2025

Uh oh!

pytorch-bot bot commented Nov 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15963

❌ 6 New Failures

Uh oh!

meta-cla bot commented Nov 24, 2025

Action Required

Process

Uh oh!

github-actions bot commented Nov 24, 2025

This PR needs a release notes: label

Uh oh!

meta-cla bot commented Nov 24, 2025

Uh oh!

metascroy commented Nov 24, 2025

Uh oh!

Lokesh9106 commented Nov 24, 2025 via email

Uh oh!

metascroy commented Nov 24, 2025

Uh oh!

metascroy commented Nov 25, 2025

Uh oh!

metascroy Nov 25, 2025

Choose a reason for hiding this comment

Uh oh!

metascroy Nov 25, 2025

Choose a reason for hiding this comment

Uh oh!

pytorch-bot bot commented Dec 3, 2025

Uh oh!

metascroy commented Dec 3, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

pytorch-bot bot commented Nov 24, 2025 •

edited

Loading

This PR needs a `release notes:` label